adk-evals
ADK Evals Skill
What are Evals?
Evals are automated conversation tests for ADK agents. Each eval defines a scenario — a sequence of user messages or events — and asserts on what the bot should do: what it says, which tools it calls, how state changes, which workflows run, and more.
Evals run against a live dev bot (adk dev), so they test the full stack — not mocks.
When to Use This Skill
Use this skill when the developer asks about:
- Writing evals — file format, assertions, turn types, setup
- Running evals — CLI commands, filtering, output interpretation
- Testing specific primitives — how to test actions, tools, workflows, conversations, state
- The testing loop — write → run → inspect traces → iterate
- CI integration — exit codes,
--format jsonflag, tagging strategies - Eval configuration — idleTimeout, judgePassThreshold, judgeModel
More from botpress/skills
adk
a set of guidelines to build with Botpress's Agent Development Kit (ADK) - use these whenever you're tasked with building a feature using the ADK
600adk-frontend
Guidelines for building frontend applications that integrate with Botpress ADK bots - covering authentication, type generation, client setup, and calling bot actions
239adk-integrations
guidelines for discovering, adding, configuring, and using Botpress integrations in ADK projects - use when users ask about connecting services, managing dependencies, or using integration actions
208adk-debugger
Systematic debugging for ADK agents — trace reading, log analysis, common failure diagnosis, and the debug loop.
206adk-docs
guidelines for creating, reviewing, updating, and searching ADK documentation - use when users ask about writing, maintaining, or auditing ADK bot docs
205adk-dev-console
Explains the ADK Dev Console — what each tab shows, how to read Agent Steps, traces, and other UI features visible at localhost:3001 during adk dev
166